HiveAI

A lightweight web app that fans prompts across Amazon Bedrock models (Nova, Claude, DeepSeek, Llama, Qwen, OpenAI, and Writer) and returns a combined summary.

Setup

1. Install dependencies:

   npm install

2. Create a .env file with AWS credentials and region that can invoke the desired Bedrock models:

   AWS_ACCESS_KEY_ID=your_access_key
   AWS_SECRET_ACCESS_KEY=your_secret_key
   AWS_REGION=us-east-1

   The UI enables all models only when these values are present.
3. Start the server:

   npm start

4. Open http://localhost:3000 to chat with the hive.

Providers

All providers run through Amazon Bedrock:

• Amazon Nova Lite (amazon.nova-lite-v1:0)
• Anthropic Claude 3.5 Sonnet (anthropic.claude-3-5-sonnet-20241022-v2:0)
• DeepSeek V3.1 (deepseek.r1-v1:0)
• Meta Llama 3.2 90B Instruct (meta.llama3-2-90b-instruct-v1:0)
• OpenAI GPT-OSS-120B (mapped to openai.gpt-4o-mini-2024-07-18-v1:0)
• Qwen Qwen3 235B A22B 2507 (qwen.qwen3-235b-a22b-instruct-v1:0)
• Writer Palmyra X5 (cohere.command-r-plus-v1:0)

The app auto-summarizes all provider answers into a single bullet list using Nova Lite.

Notes

• Make sure your AWS account has Bedrock access to the listed model IDs.
• Frontend calls /api/query with your prompt and chosen providers; responses show success or error messages per engine and include a unified summary panel.